/. Embryol. exp. Morph. Vol. 32, 1, pp. 111-131, 1974
Printed in Great Britain
\l\
Cellular aspects of regeneration hormone
influence in Nereis diversicolor
By P. J. W.OLIVE 1
From the Department of Zoology and Dove Marine Laboratory,
University of Newcastle upon Tyne
SUMMARY
The histology of caudal regeneration in Nereis diversicolor is described with special
reference to the role of the cerebral 'regeneration' hormone. Wound healing is complete
after 8 days; the regenerated stump of tissue then consists of the pygidium, associated anal
cirri and a prepygidial region between the pygidium and the posterior margin of the last intact
segment. The term segment blastema is used to refer to the cells from which all segmental
structures are derived. The segment blastema is situated in the ventral posterior margin of
the prepygidial region, and consists of transverse bands of cells with large nuclei (12 fim
diameter) and prominent nucleoli. Segment anlagen are formed immediately in front of the
segment blastema; the development of the anlagen is described, and six stages of development are defined.
The extent of caudal regeneration in different animals is best compared with reference
to the total number and stage of development reached by the segment anlagen. During
normal regeneration at 18 °C, segment anlagen are first formed after 8 days, and they
continue to be formed at the same rate until at least the 21st day after caudal ablation. The
oldest segments reach stage 6 after about 15 days of regeneration.
Wound healing, pygidium formation, and cirrus development and establishment of the
segment blastema all occur in decerebrate animals following the loss of caudal segments, but
segment anlagen formation is almost completely inhibited. Implantation of ganglia taken
from intact donor animals into the coelom of decerebrate animals which have lost caudal
segments initiates segment anlagen formation. The segment blastema of decerebrate animals
remains competent to respond in this way for at least 15 days.
The regeneration hormone produced by the cerebral ganglion is essential for continued
segment anlagen production throughout the 2nd and 3rd weeks of regeneration at 18 °C.
Delayed decerebration leads to an arrest of anlagen production, and anlagen younger than
stage 3 fail to develop further. Older anlagen (stages 4 and 5) are independent of the regeneration hormone and can continue to differentiate in its absence.
INTRODUCTION
Nereis diversicolor, in common with most polychaete annelids, can regenerate
lost caudal segments; however, they are only able to do so in the presence of a
hormone released by the cerebral ganglion (Durchon, 1956; Clark & Bonney,
1960; Durchon & Marcel, 1962) and the secretion of this hormone ceases in
1
Author's address: Dove Marine Laboratory, Cullercoats Bay, North Shields, Northumberland, NE3O 4PZ, U.K.
112
P. J. W. OLIVE
sexually mature animals, which are therefore not able to regenerate (Golding,
1967e; Porchet & Durchon, 1968).
The role of the cerebral hormone in the process of regeneration, however,
is unknown, although much experimental work has been directed towards
discovering whether the hormone has a triggering effect on regeneration in
response to segment loss (Clark & Bonney, 1960; Clark & Evans, 1961;
Durchon & Marcel, 1962; Clark & Ruston, 1963; Clark & Scully, 1964;
Scully, 1964) or if the hormone has a prolonged effect on regeneration, rather in
the manner of a growth hormone (Golding, 1961 a-d). Recent reviews have
considered the latter hypothesis to be the more likely (Clark, 1965; Golding,
1967e; Clark & Olive, 1973).
At the cellular level, Herlant-Meewis & Nokin (1963) and Boilly (1965, 1969)
have made detailed descriptions of wound healing and tissue activation, but most
of their observations relate to events occurring during the first few days of regeneration. Herlant-Meewis & Nokin (1963) thought that either the accumulation
of coelomocytes to form the wound plug, or tissue activation, both of which occur
during the first 3-4 days of regeneration, might be processes which only occurred
in the presence of the regeneration hormone. However, since that time, experimental evidence has emphasized the long-term influence of the regeneration
hormone, and a re-examination of the possible role of the hormone at the tissue
level is essential. The present paper, which describes the histology of regeneration in relation to experimental investigation of the effect of the cerebral hormone, is the first part of such a study. A report of an investigation into the
influence of the regeneration hormone on the kinetics of cell division in the
regenerating tissue will follow.
MATERIALS AND METHODS
Routine maintenance
Animals were collected from the river Wansbeck estuary near Blyth, Northumberland. They were maintained in the laboratory in filtered 100% sea water
at 16-18 °C. Experimental conditions and treatment were based on those
described by Golding (1967c, d) in similar work. The animals were kept in
groups of up to 30 in plastic containers with aeration and an abundant supply
of glass tubes. Where animals were kept together without decerebration the
jaws were removed after narcotization with MS 222. Ganglion extirpation was
performed using fine scissors, the posterior two-thirds of the ganglion being
removed, together with overlying epidermis and associated glandular structures.
The palps, tentacles and cirri were left intact. Caudal segments were removed by
autotomy (provoked by tightly grasping the animals with forceps at the 25th
segment) at the 25th-28th segment from the pygidium. Experimental animals
had between 65-80 segments and were not sexually mature. (Females with
oocytes greater in diameter than 120 ftm were excluded.)
Hormone influence on Nereis regeneration
113
In some experiments, ganglia were implanted dorso-laterally through the
body wall into the coelom of a previously decerebrated host animal. Donor
animals for such implantations were similar to experimental animals and did
not have caudal segments removed prior to decerebration. Apart from the use
of filtered sea water, which was changed every 2nd or 3rd day, no further
attempt to produce sterile conditions was made.
Histological techniques
All experimental animals were narcoticized at the termination of the experiment and the last few segments plus regenerated tissue were fixed in sea-water
Bouin. The material was paraffin embedded, cut at 6 /*m in frontal or sagittal
planes and stained with azan trichrome stain.
RESULTS
Morphological and histological observations
Segmentation during caudal regeneration in Nereis diversicolor begins after
the completion of wound healing (Herlant-Meewis & Nokin, 1963), which
occurs at 18 °C about 8 days after caudal ablation. By this time the components
of the regeneration system are fully established, and two main regions can
be distinguished: the pygidium and associated structures, and the prepygidial
region, in the floor of which lies the segment blastema.
The structure of the pygidium after wound healing
The pygidium is separated from the more anterior parts of the regenerate by a
muscular septum carrying a large blood vessel, which connects the dorsal and
ventral blood vessels running into the regenerating stump from the previous
segment. Two extensions of the damaged nerve cord extend into the newly
formed pygidium; they enter mid-ventrally, pass diagonally across the pygidial
floor and enter the anal cirri, which by 10 days are well developed. The cirri are
solid, with an axial neuropile; mitoses are frequently seen in them, and at their
base in the floor of the pygidium. This is thickened latero-ventrally to form the
pygidial cushions, which contain two types of secretory material. In the surface
epithelium there are goblet cells which stain deep red with azan, while the cells
of the pygidial cushions are interspersed with coiled capsules containing a
finely granular substance staining pale blue with azan.
The space within the pygidium surrounding the opening of the gut via the
anus becomes lined with muscle cells. Occasionally, large numbers of coelomocytes remain within the pygidium, but they have usually dispersed prior to
segmentation.
E M B 32
114
P. J. W. OLIVE
°©£? / \ Neuropile
Anal cirrus
segment
anlagen
,
50/*m
Stage 2
segment
anlagen
Septum
blood vessel
Developing fibroblasts
Chaetal sac complex
Fig. 1. Semi-diagrammatic drawing of part of a frontal section at the level of the
ventral blood vessel through the regenerating tail of N. diversicolor 12 days after
caudal ablation. The anal cirri, pygidium, segment blastema and five segment
anlagen (I-V) are present. Segment V has a completed septum with developing
blood vessel and chaetal sac complex at stage 2. Segment IV shows the separation
of the chaetal sac complex from the developing fibroblasts which will form the
septum. In segment III this separation is not complete. The two youngest segment
anlagen (I and II) are at stage 1.
The structure of the prepygidial region and the location of the segment blastema
The prepygidial region consists of a truncated cone of tissue between the
pygidial septum and the muscular septum at the posterior edge of the last
complete segment. The gut projects through this cone without interruption.
Surrounding the gut is a ring-shaped coelomic cavity which is at first asegmental.
Prior to the growth of the segment anlagen, the greater part of this region is
derived from the pre-existing tissues of the damaged segment and consists of
modified epidermis, muscle cells, fibroblasts and coelomocytes. The ventral
floor, however, is dramatically different and it is here that the segment anlagen
are formed. Mid-ventrally, in the epidermis, there are two neuropile tracts
which lead into the pygidium and on to the anal cirri. Within the prepygidial
region the neuropile is surrounded by small, densely staining cells which are
mitotically active. These do not accompany the neuropile tracts into the pygidium. At the boundary between the prepygidial zone and the pygidium, two
small nerves pass outwards. Immediately in front of these lies the segment
blastema (Figs. 1, 2). The blastema is characterized by large cells with spherical
nuclei about 10-12 jam in diameter with prominent, deeply staining nucleoli
about 5 jLcm in diameter. These large cells can be recognized both in the ectoderm and in the mesoderm; in the latter, being interrupted by the ventral nerve
cord extensions and the ventral blood vessels, they form two bands of prominent cells described as the 'bandelettes mesodermique' by Boilly (1969). The
Hormone influence on Nereis regeneration
115
ectoderm of the segment blastema lies below the mesoderm, and is a somewhat
broader zone which merges posteriorly with similar cells in the ventral floor of
the pygidium. The latter, however, are smaller (diameter 3-8 /im, Fig. 1).
Hofmann (1966), describing regeneration in Platynereis, also recognized the
importance of these cells, which he termed the ' embryonalzellen der Proliferationszone\
It must be emphasized that the blastema cells are not the only cells which
are dividing by mitosis, but they are thought to play a stem cell role, maintaining
a source of undifferentiated cells throughout regeneration, and indeed, throughout normal growth.
The development and differentiation of segment anlagen
During caudal regeneration, several segment anlagen form, which will
differentiate into fully developed segments. In any regenerating individual the
oldest segments are most advanced and are situated farthest from the segment
blastema, while the youngest are in contact with it.
In order that comparisons could be made rapidly and accurately between
experimental animals, it was found necessary to develop a criterion of regeneration other than merely the number of segment anlagen, as used by previous
authors. The following sequence of stages was developed to describe the
development of an individual segment anlagen. As will be seen below, in a
normally regenerating specimen there will be a number of segment anlagen
developing simultaneously, with perhaps several in each of these stages.
Stage 1. The earliest indication of the onset of segmentation is the appearance
of lateral segmental nerves from the ventral nerve cord extensions. This is very
soon followed by the onset of segmentation in the mesoderm. Some of the
mesoblasts derived from the segment blastema begin to differentiate as fibroblasts which form the septal anlagen. As they do so, they extend dorsally to
contact the ventral blood vessel and eventually the gut, so that the asegmented
cavity of the prepygidial zone becomes segmented by the dorsal growth of the
septal anlagen. The fibroblast-like differentiation does not affect all the mesoblasts, and between successive septal anlagen are masses of mitotically active
cells which will form the mesodermal component of the chaetal sac complex
(Figs. 2, 3).
At the end of wound healing, the prepygidial region is already between 50
and 150 jtim in length, and as many as three segment anlagen may be established
almost simultaneously (Fig. 2). Subsequent anlagen arise by the differentiation
of transverse bands of mesoblasts as septal anlagen, with chaetal sac anlagen
between.
A space begins to develop which separates the chaetal sac anlagen from the
septal anlagen and which will become the parapodial coelom (Fig. 3). Mesoderm
segmentation proceeds as if the underlying ectoderm were growing more rapidly,
causing successive rows of mesodermal cells to be separated from each other.
8-2
116
P. J. W. OLIVE
Endoderm
Anal
cirrus
Neuropile
Pygidial
septum
Ventral
blood vessel
Segment
blastema
Invaginating
ectoderm forming the
1st chaetal sac
Fig. 2. Semi-diagrammatic drawing of a parasagittal section through the regenerating
tail of N. diversicolor 8 days after caudal ablation. Wound healing is complete;
anal cirri, pygidium and segment blastema have formed and one segment anlagen is
present, with ectoderm invaginating to form the first chaetal sac anlagen.
Stage 2. Stage 2 can be defined by the appearance of the ectodermal component of the parapodial/chaetal sac anlagen. This is formed by the invagination
of ectodermal cells into the mesoderm cell group lying between the septal
anlagen. Only small numbers of cells are invaginated, through finger-like
intuckings of the ectoderm (Fig. 2), but the invagination invariably includes
some large cells which resemble those of the blastema and which ultimately
become the aciculoblasts and chaetoblasts (see also Figs. 4-6).
In frontal sections, the ectodermal and mesodermal components of the
chaetal sac can be recognized by the basement membrane (Fig. 3). Mitosis is
common in both the ectoderm and mesoderm cells of the chaetal sac.
During stage 2 the septal anlagen continue to develop, and may eventually
contact the gut and even the inner dorsal surface of the animal. Simultaneously,
growth of the ectoderm carries the chaetal sac anlagen laterally and a prominent
segmental ridge may appear.
Stage 3. Stage 3 can be defined by the appearance of the notopodial aciculum.
One of the large invaginated ectoderm cells in the chaetal sac becomes flattened
on one surface and begins to secrete the aciculum (Fig. 3). This aciculoblast
remains at the base of the developing aciculum (Figs. 3-5). During stage 3
segmentation of the prepygidial zone is completed, and the septal anlagen meet
at the dorsal surface. The blood system may also become segmented at this
117
Hormone influence on Nereis regeneration
Anlagen
Anlagen
Blastema
Anlagen
III
Stage 3
Pygidial
septum
Developing
segmental
blood vessel
Neuropile
Aciculoblast
Aciculum
Mesoderm
and
ectoderm
in the
segment
blastema
Secretory
material
in pygidium
Basement
membrane
Developing
coelom
Invaginating
ectoderm of the
chaetal sac
50//m
Fig. 3. Camera lucida drawing of part of a frontal section through the regenerating
tail of N. diversicolor 21 days after caudal ablation. The drawing shows the segment
blastema and 3 segment anlagen at stages 1, 2 and 3.
Segmental
blood vessel
Muscle cells
,,.L .
Mitosis
—«—,..—, ,^,™, , ^ . .
Aciculoblast
Aciculum
Chaetoblast cells
Developing
dorsal cirrus
Fig. 4. Camera lucida drawing of a frontal section through the parapodium of a
stage 4 segment anlagen in N. diversicolor, 21 days after caudal ablation.
118
P. J. W. OLIVE
I
50/i m
Coelom
Muscle
cells
Muscle
cells
Aciculoblast
Chaetoblast
cells
Notopodial
aciculum
Neuropodial
aciculum
Chaetae
Coelom
Fig. 5. Camera lucida drawing of a frontal section through the parapodium of
stage 5 segment anlagen in N. diversicolor, 21 days after caudal ablation.
time, with the appearance of a segmental blood vessel in the septum. The
essential components of the parapodial complex have been established at this
stage, and subsequent stages of the development are based on arbitrarily chosen
criteria, which have nevertheless proved useful for the purpose of comparisons
between individual animals.
Stage 4. Stage 4 is defined by the appearance of the dorsal cirrus of the
parapodium and the further growth of the aciculum (Fig. 4). The acicular sac
elongates and projects into the now spacious coelom.
Stage 5. Stage 5 is defined by the appearance of the neuropodial aciculum.
Differentiation of the notopodium is well advanced. Muscle cells begin to
differentiate from the mesodermal component of the chaetal sac and the secretion of the chaetae is initiated, though these do not yet project from the chaetal
sac (Fig. 5).
Stage 6. Stage 6 is defined by extrusion of the notopodial chaetae through
the epidermis (Fig. 6). This represents the terminal event of this staging sequence,
though clearly the parapodial anlagen are far from fully developed. At stage 6
the segments have septa with blood vessels, a spacious coelom, and parapodia with dorsal and ventral aciculae, dorsal and ventral cirri and at least
119
Hormone influence on Nereis regeneration
Aciculoblast
Chaetoblast cells
Coelom
Mesoderm
Ectoderm
Fig. 6. Camera lucida drawing of a frontal section through the parapodium of a
stage 6 segment anlagen in N. diversicolor, 21 days after caudal ablation.
Table 1. Staging systems {designation and definition) for parapodial development
in Nereis diversicolor and Platynereis dumerilii
N. diversicolor
(Olive)
Stage 1. Lateral nerves appear. Septae begin
to develop in mesoderm
Stage 2. Ectoderm invagination to produce a
chaetal sac anlage
Stage 3. Notopodial aciculum secretion
begins
Stage 4. Appearance of dorsal cirrus.
Extrusion of aciculum
Stage 5. Appearance of neuropodial aciculum. Appearance of chaetae
Stage 6. Extrusion of chaetae. Dorsal and
ventral cirri
PL dumerilii
(Hoffmann, 1966)
Stage I. Parapodial anlage, an undivided
cup with or without chaetae
Stage II. Parapodium with notopodium and
neuropodium, but no cirri
Stage III. Parapodium with notopodium
and neuropodium, dorsal cirrus and
chaetae
Stage IV. Parapodium with notopodium,
neuropodium, chaetae, dorsal and ventral
cirri
some functional chaetae. The intrinsic parapodial musculature is also well
developed.
In developing this sequence of staging, the appearance of the segmental blood
vessels was specifically avoided as a staging criterion, partly because this event is
variable with respect to the appearance of the parapodial rudiments, and also because in decerebration experiments to be described below, some segment anlagen
became arrested at stage 2 but did nevertheless acquire a segmental blood vessel.
120
P. J. W. OLIVE
(30)
i
)
•5
a 7g
6-
|
5-
°
4-
S
3-
a
a
2-
3
(2 1 )
(16)
(34)
\
1 '
(22)
03
S
'
J
1-
1
2
4
I
I
I
I
I
I
I
I
6 8 10 12 14 16 18 20 22
Duration of regeneration (days)
Fig. 7. The number of segment anlagen regenerated during caudal regeneration of
Nereis diversicolor at 18 °C. Vertical bars give the 95 % confidence limits of the
mean. The figure in brackets at each point is the number of animals contributing
to the mean.
The most detailed comparable staging system for the ageing of regenerating
segments in Nereids is that proposed by Hofmann (1966) for PI. dumerilii.
The two systems are compared in Table 1. The main differences between the
two schemes is in the timing of the appearance of the chaetae, which occurs
much earlier in Platynereis than in Nereis.
Experimental analysis of cerebral hormone function in regeneration
Development of segmental anlagen in normal regeneration
There have often been marked differences in the rate of regeneration recorded
for Nereis diversicolor by different authors. It is therefore essential to define for
any sequence of experiments the 'normal' regeneration in the conditions
adopted. This is shown in Fig. 7 for results from 123 animals. The number of
segment anlagen was determined from histological sections and therefore includes
segment anlagen not visible externally. These results agree closely with those of
Golding (1967c?), which also indicated that the majority of segments appear
during the 2nd and 3rd weeks of regeneration. Figure 8 shows the development
of the segment anlagen during the first 21 days of regeneration. The first segments
reach stage 6 (of the developmental sequence defined above) after about 15
days, and this represents the most numerous class of segment anlagen after
21 days of caudal regeneration. Segment delineation (the appearance of the
youngest anlage) starts at about the 8th-10th day following caudal ablation,
and continues throughout the 2nd and 3rd weeks, so that stage 1 anlagen are
still present at an average of about 1 per individual on day 21.
Hormone influence on Nereis regeneration
Stage
1 2
i
Days of
regeneration
I I
3
4
i
5
i
121
6
i
100-5-
10
12
15
21
Fig. 8. Diagram showing the mean number of segments at each stage of development on the 8th, 10th, 12th, 15th and 21st day after caudal ablation of Nereis
diversicolor.
Regeneration in decerebrate animals
It is now well established that after removal of the supra-oesophageal ganglion, Nereis diversicolor loses the ability for caudal regeneration. This has been
confirmed in the present investigation. However, sections of the posterior parts
of decerebrate animals that have lost posterior segments show that there is not a
complete failure of all features of wound healing and regeneration.
In the majority of individuals, wound healing and pygidium formation
proceed almost normally, to such an extent that the anal cirri, a pygidium with
anal sphincter muscles, prepygidial zone and segment blastema cells can be
recognized. In approximately 25 % of decerebrate animals one or other of these
122
P. J. W. OLIVE
Total stage
1
,
2
.
3
,
4
,
5
.
6
.
lO-i
0-5 •
10
10
a
o
i
0-5
00
u
ber
00
<u
"o
10
S 0-5
3
en
C
Q 17
1-5
21
0-5
Fig. 9. The mean number of segments and their stage distribution in caudal
regeneration of decerebrate Nereis diversicolor. The 95 % confidence limits for the
mean number of segments are indicated by vertical bars. The means are not
significantly different.
features is missing. A similar analysis of all animals with the brain intact reveals
these structures to be present in 95 % of cases. In many decerebrate animals,
segment anlagen are found in the prepygidial zone. Of the 89 decerebrate
animals that have been examined, 33 (37 %) have at least 1 segmental anlage,
and a small number have three recognizable anlagen.
Figure 9 shows the mean number of segment anlagen and the age distribution
of anlagen in decerebrate animals. The regenerating stumps remain arrested at a
stage more or less equivalent to that in normal animals at 8 days regeneration,
when wound healing is complete. The few segment anlagen that are produced
fail to differentiate further.
Regeneration after delayed implantation of supra-oesophageal ganglia
Golding (1974) has shown that decerebrate Nereis diversicolor can be made to
regenerate new segments after the completion of wound healing in the absence
of the brain, by the implantation of ganglia into the coelom.
Similar experiments have been designed for observation of the histology of
caudal regeneration following delayed implantation. Animals were allowed to
regenerate for 10 or 15 days prior to implantation and were examined 3, 7 and
15 days later. The mean number of segment anlagen was significantly increased
7 days after implantation, i.e. about 2 days earlier than could be detected by
external observation (Golding, 1974). None of the anlagen present 7 days after
Hormone influence on Nereis regeneration
123
Table 2. The mean number of segment anlagen, ± standard deviation,
regenerated by Nereis diversicolor after delayed decerebration
(Number of animals in parentheses.)
Days after decerebration
Days of regeneration
prior to decerebration
5
10
15
20
r
5
2 •2±1 •5 (6)
2 0 ± l •3 (8)
4 •8 + 1•6 (5)
8 •3 + 2•1 (3)
10
20
1-75 ±0' 9 (4)
1- 5 ±1-•3 (9)
6 2 + 1 6 (5)
6 •5 ±2 •1 (2)
1-7 i l - 3 (4)
3 (9)
6-711- 3 (4)
6-312- 1 (3)
decerebration had progressed beyond stage 2, but 15 days after implantation
the most advanced were at stage 6 and the distribution of segment anlagen was
similar to that occurring after 15 days of normal regeneration (Fig. 8).
At the time of implantation, the processes of wound healing and segment
blastema formation were complete. Nevertheless, there was a delay of 7 days
prior to the appearance of new anlagen, and the rate at which they differentiated
was very similar to that following caudal ablation of animals with the cerebral
ganglia in situ.
The effect of delayed decerebration on regeneration
In these experiments the animals were left for various times after caudal
ablation before the supra-oesophageal ganglia were removed. In the first
experiment of this kind, four groups of animals were allowed to regenerate for
5, 10, 15 and 20 days prior to decerebration. They were left for a further 5, 10,
and 20 days before fixation and preparation for histological examination. The
mean number of segment anlagen regenerated and the age distribution of these
was determined. The results are shown in Table 2 and Fig. 10, but these should
also be compared with the results for normal regeneration shown in Figs. 7 and
8. Those animals decerebrated on or before the 10th day of regeneration, when
segment anlagen production is just beginning, are arrested at this stage of
regeneration. The number of segment anlagen produced remains low, and those
anlagen that are present are arrested in early stages of development (prior to
stage 3 in almost all cases). Those animals that were decerebrated 10 days after
caudal ablation, and allowed to regenerate for a further 10 days, have regenerated for a similar period of time to those in Fig. 8 (21 days) but are obviously
very different from them.
The animals decerebrated at the 15th day are particularly interesting. At that
time more anlagen are established (mean number = 4-5) and of these a small
number may have reached stages 5 and 6 of the developmental sequence. Figure 10
124
P. J. W. OLIVE
Days regeneration
post decerebration
5
,
10
Stage of anlagen development
Fig. 10. Stage distribution of segment anlagen regenerated after delayed
decerebration. Details of experimental treatment in text.
shows that the more advanced of the segment anlagen at the time of decerebration are able to continue to develop in the absence of the brain. However,
further production of new segments appears to decline, and the younger segment anlagen remain in stages 1-3. The results for animals decerebrated on the
20th day are similar.
A second experiment was designed to investigate in greater detail and with
more animals the role of the supra-oesophageal ganglion during the critical
period of segment production between the 8th and 12th days of caudal regeneration.
The experimental design consisted of the following groups:
(i) Brain intact during 21 days regeneration (code: 21/—).
(ii) Decerebrate throughout 21 days regeneration (0/21).
(iii) Decerebrated after 8 days regeneration, examined after 21 days of regeneration (8/13).
(iv) Decerebrated after 10 days regeneration, examined after 21 days of
regeneration (10/11).
Hormone influence on Nereis regeneration
Treatment code 0/21
125
21/—
6
5
4
3
1 2
8/-
10/-
12/-
8/13
10/11
12/9
3o
X>
1c
a
«
2
54
321-
Fig. 11. The mean number of segments regenerated during caudal regeneration in
Nereis diversicolor with delayed decerebration. Vertical bars show the 95 %
confidence limits for the means. Explanation of treatment code: 0/21, Decerebrate
during 21 days regeneration; 2.1/—, brain intact during 21 days regeneration;
8/—, brain intact during 8 days regeneration; 10/—, brain intact during 10 days
regeneration; 12/—, brain intact during 12 days regeneration; 8/13, brain intact
during 8 days regeneration, fixed after a further 13 days of decerebrate regeneration; 10/11, brain intact during 10 days regeneration, fixed after a further 11 days
regeneration; 12/9, brain intact during 12 days regeneration, fixed after a further
9 days regeneration.
(v) Decerebrated after 12 days regeneration, examined after 21 days of
regeneration (12/9).
(vi, vii, viii) Fixed and examined after 8, 10 and 12 days of caudal regeneration respectively, with the brain intact (8/—, 10/—, 12/—).
The number of segment anlagen and the stage of their development was
determined from histological sections. Figure 11 summarizes the mean number of
anlagen in each group; an analysis of variance for all the animals fixed on the
21st day showed that animals which regenerated with the brain intact for 21
126
P . J. W . O L I V E
21/-
12/-
111 2 1 31 4 1 5| 6 |
10/11
12/9
2 | 3| 4 | 5| 6|
I 2| 3 | 4 | 5 | 6 |
Stage of segment anlagen development
Fig. 12. Age distribution of segment anlagen regenerated during caudal regeneration
in Nereis diversicolor with delayed decerebration. See text and legend to Fig. 11 for
details of experimental treatment of each group.
days had significantly more segment anlagen than all the other groups, including
animals which regenerated for 12 days with the brain intact (group 12/9). Decerebration clearly leads to an arrest in the process of segment anlagen production,
although a small number of anlagen may be produced after decerebration.
There are more segments in all groups of animals that have regenerated for 21
days (8/13, 10/11, 12/9) than there are in the corresponding control groups
(8/—, 10/—, 12/—) but the differences are small, and in the case of the animals
decerebrate or fixed on the 10th day (groups 10/11 and 10/—), not significant.
There is therefore only a limited capacity for further anlagen production after
decerebration.
The stage distribution of the segment anlagen for each group is shown in
Fig. 12. The eldest segments present at the time of decerebration continue to
develop normally and may reach stages 5 and 6. The number of such segments
is greater the longer the delay prior to decerebration. Most of the segment
anlagen which were at early stages of segment anlagen at the time of decerebration remain arrested at that stage. A few of the stage 2 anlagen may acquire
a small centre of aciculum secretion, but they remain small and do not develop
further.
Hormone influence on Nereis regeneration
127
The appearance of the younger segments suggests that their growth is arrested
by decerebration, and that before a certain size has been reached the segment
anlagen are not able to differentiate and develop in the absence of the brain.
DISCUSSION
Morphological and histological studies
Caudal regeneration in Nereis diversicolor resembles closely the same process in Platynereis dumerilii (Hofmann, 1966) and Nephtys (Clark & Clark, 1962;
Clark, 1965, 1968). In all three species, wound healing and pygidium formation
are separate processes from, and largely independent of, the formation of new
segments. New segments arise from the products of cell division in a narrow
band of specialized cells lying ventrally immediately anterior to the pygidium.
The cells in this region have a very striking appearance, with a large nucleus
which appears empty apart from the prominent, deeply staining nucleolus.
In Nereis and Platynereis (Hofmann, 1966) cells of this type are found in both
the ectoderm and the mesoderm in the prepygidial region at the junction with
the pygidium, but in Nephtys (Clark, 1968) the ectodermal cells lie more
posteriorly, in the floor of the pygidum. These cells have been recognized by a
number of authors - Herlant-Meewis & Nokin (1963), Hofmann (1966),
Clark (1968) and Boilly (1969) - but the terminology used for these cells and
other structures in regeneration has not been consistent.
Clark & Clark (1962) and Clark (1968) have used the term 'blastema' to
include all the cells which accumulate at the wound, although Clark (1968)
points out that this may not be ideal, since only a small proportion of these
cells actually participate in the regeneration of new tissue. The general term
'wound plug' equivalent to the French 'bouchon cicatricielle' (HerlantMeewis & Nokin, 1963; Boilly, 1969) is preferable for this accumulation of
cells.
In Nephtys (Clark, 1968), Nereis diversicolor (Boilly, 1969) and Platynereis
(Hofmann, 1966) the cells which give rise to the segmental mesoderm have been
identified on the anterior ventral face of the newly regenerated pygidium. These
cells have been described as the 'zone of proliferation' (Clark, 1968) in Nephtys,
the 'bandelettes mesodermique' in Nereis diversicolor (Boilly, 1969) and 'die
embryonalzellen der Proliferationszone' in Platynereis (Hofmann, 1966). As
these cells are thought to be the ultimate source of all the segmental structures
(Hofmann, 1966; Clark, 1968), the term 'segment blastema' may be applied
to them. This term is equivalent to the term 'proliferation zone', but has been
preferred in the present report because mitosis and cellular proliferation are
equally, if not more, common in the pygidium and anal cirri, in the younger
segmental anlagen and amongst the cells of the ventral nerve cord.
128
P. J. W. OLIVE
Cerebral hormone dependence in caudal regeneration in
Nereis diversicolor
Earlier work on the experimental analysis of regeneration and the role of the
supra-oesophageal ganglion in Nereis diversicolor has been mainly concerned
with external features. However, several authors (Clark & Bonney, 1960;
Clark & Evans, 1961; Durchon & Marcel, 1962; Golding, 1967c) have noticed
that a pygidium will form in the absence of the supra-oesophageal ganglion.
The histological part of the present investigation has confirmed that this is the
case. Perhaps even more importantly, it has shown that the cells which comprise
the segment blastema are formed and appear in the normal position in at least
75 % of decerebrate animals. The regeneration hormone is not therefore essential for the establishment of the segment-forming tissues. Instead, it appears
that the cerebral hormone is necessary for the proper functioning of the
blastema, and also for the growth of the young segment anlagen. This is illustrated most clearly by the delayed decerebration experiment, described in
Table 2 and in Figs. 10-12. These experiments have shown that when decerebration takes place at a time when new segment anlagen are being formed,
segment production soon stops and the youngest segments are arrested at early
stages of differentiation. The older segments, and especially those that have
proceeded to stage 4, in which the main components of the parapodium are
present, are able to continue to differentiate in the absence of regeneration
hormone.
In Nereis, as in Nephtys (Clark, 1965, 1968), the early stages of segment
differentiation are characterized by a high nucleus/cytoplasm ratio and the
developing cells are mitotically very active. The later stages of development,
however, involve a gradual change to growth by cell differentiation and in the
latter phase the regeneration hormone does not seem to be involved.
In the light of these observations it seems most likely that the cerebral hormone has a mitogenic effect. However, absence of the supra-oesophageal
ganglion does not lead to a general failure of mitosis, as is shown by the growth
of the anal cirri. Furthermore, preliminary work has suggested that decerebration does not lead to an immediate and complete failure of DNA synthesis in the
segmented blastema, nor in the young segment anlagen (P. J. W. Olive,
unpublished observations); this aspect of the effect of the regeneration hormone is being further investigated.
The central nervous system of Nereis diversicolor has two separate effects on
regeneration; in addition to the endocrine influence of the cerebral ganglion,
normal regeneration requires the presence of nervous tissue derived from the
cut ends of the ventral nerve cord. Without this regeneration can occur, but it
is abnormal; parapodia and anal cirri do not form, though other regeneration
events occur normally (Boilly & Combaz, 1970; Combaz & Boilly, 1971;
Combaz, 1972). The nervous system is necessary for anal cirrus development and
Hormone influence on Nereis regeneration
129
for the inductive events which lead to the establishment of parapodia (BoillyMarer, 1971a, b). On the other hand, the main influence of the cerebral
ganglion appears to be on the establishment of the mesodermal components of
the segments and their subsequent growth, without which proper differentiation
of the segments cannot take place.
Evidence in support of the hypothesis that the regeneration hormone is a growth
hormone
The delayed implantation and delayed decerebration experiments establish
that the cerebral hormone is responsible for the appearance of segment anlagen
and their early growth, but that eventually the segment anlagen become
independent and can continue to differentiate in the absence of the hormone.
Segment production does not begin until the 8th day at 18 °C and is then very
active throughout the next 2 weeks. Normal regeneration will only take place
if there is a source of cerebral hormone throughout that period. Several investigators have used the delayed decerebration technique (Clark & Evans, 1961;
Clark & Ruston, 1963; Hofmann, 1966; Golding, 1967 d) and all have found that
the number of segments regenerated increased with the length of time that the
brain remained in situ. There have, however, been differences in interpretation
of the results (see Golding 1967a, c, d) which have arisen at least in part from
the failure to adequately define the criteria used to define a segment. The staging
sequence adopted here avoids this difficulty and confirms that the cerebral
ganglion has a prolonged effect on segment formation.
If the later stages of segment anlagen development such as the appearance
of chaetae were used to define 'a segment' then it would appear from external
morphological evidence that segments continued to appear long after decerebration - which might be construed as evidence for a 'trigger-like' effect on
regeneration. In reality there is no such effect. In the implantation experiments
it was found that ganglia implanted in decerebrate hosts which had completed
wound healing would initiate segment formation and development. It was not
necessary for the ganglia to be 'activated' by caudal ablation of donor animals
as suggested by Scully (1964), and in all cases the ganglia were taken from
animals which had not lost caudal segments.
Caudal regeneration is not under the influence of the supra-oesophageal
ganglion in all species of polychaete (see Hill, 1972), and in fact, this dependence
has only been demonstrated in members of the families Nereidae and Nephtydae. In the Nereidae, but not the Nephtydae (P. J. W. Olive, unpublished
observations), the life-history is divided into a period of somatic growth
followed by a period of sexual development. The change is mediated hormonally
(see Clark & Olive, 1973) and the hormone dependence of regeneration in Nereis
diversicolor may be one aspect of this type of endocrine regime.
E M B 32
130
P. J. W. OLIVE
REFERENCES
B. (1965). Localisation des cellules de regeneration chez Nereis diversico/or. O. F.
Muller (Annelide Polychete). C. /-. hebd. Seanc. Acad. Sci., Paris 261, 2009-2011.
BOILLY, B. (1969). Origine des cellules regeneratrices chez Nereis diversicolor O. F. Muller.
Wilhelm Roux Arch. EntwMech. Org. 162, 286-305.
BOILLY, B. & COMBAZ, A. (1970). Influence de la chaine nerveuse sur la regeneration caudale
de Nereis diversicolor (O. F. Muller). C. r. hebd. Seanc. Acad. Sci., Paris 271, 92-95.
BOILLY-MARER, Y. (1971 a). Neoformation de parapodes surnumeraires par greffe heterologue
de parois de corps chez Nereis pelagica L. (Annelide Polychete). C. r. hebd. Seanc. Acad.
Sci., Paris 111, 79-82.
BOILLY-MARER, Y. (19716). Role du systeme nerveux parapodial dans l'induction de parapodes surnumeraires par greffes heterologues chez Nereis pelagica L. C. r. hebd. Seanc.
Acad. Sci., Paris 111, 261-264.
CLARK, R. B. & OLIVE, P. J. W. (1973). Recent advances in polychaete endocrinology and
reproductive biology. Oceanogr. & mar. Biol. 11, 175-222.
CLARK, M. E. (1965). Cellular aspects of regeneration in the polychaete Nephtys. In Regeneration in Animals and Related Problems (ed. V. Kiortsis & H. A. L. Trampusch), pp. 240249. Amsterdam: North Holland Press.
CLARK, M. E. (1968). Late stages of regeneration in the polychaete Nephtys. J. Morph.
124,483-510.
CLARK, M. E. & CLARK, R. B. (1962). Growth and regeneration in Nephtys. Zool. Jb.
(Zool. und Physiol.) 70, 24-90.
CLARK, R. B. (1965). The integrative action of a worm's brain. Symp. Soc. exp. Biol. 20,
345-379.
CLARK, R. B. & BONNEY, D. G. (1960). Influence of the supra-oesophageal ganglion on
posterior regeneration in Nereis diversicolor. J. Embryol. exp. Morph. 8, 112-118.
CLARK, R. B. & EVANS, S. M. (1961). The effect of delayed brain extirpation and replacement on caudal regeneration in Nereis diversicolor. J. Embryol. exp. Morph. 9, 97-105.
CLARK, R. B. & RUSTON, R. J. G. (1963). Time of release and action of a hormone influencing
regeneration in the polychaete Nereis diversicolor. Gen. comp. Endocr. 3, 542-553.
CLARK, R. B. & SCULLY, U. (1964). Hormonal control of growth in Nereis diversicolor.
Gen. comp. Endocr. 4, 82-90.
COMBAZ, A. (1972). Potentialites regeneratrice des regenerats a neurogeniques de Nereis
diversicolor L. F. Muller. C. r. hebd. Seanc. Acad. Sci., Paris 274, 1509-1512.
COMBAZ, A. & BOILLY, D. (1971). Particularites de la morphogenese parapodial dans des
regenerats faiblement innerves chez Nereis diversicolor O. F. Muller. C. r. hebd. Seanc.
Acad. Sci., Paris 273, 951-954.
DURCHON, M. (1956). Influence du cerveau sur les processus de regeneration caudale chez
les Nereidiennes (Annelides Polychetes). Archs Zool. exp. gen. 94, 1-9.
DURCHON, M. & MARCEL, R. (1962). Influence du cerveau sur la regeneration posterieure
chez Nereis diversicolor. C. r. Seanc. Soc. Biol. 156, 661-663.
GOLDING, D. W. (1967a). Regeneration and growth control in Nereis. I. Growth and
regeneration. /. Embryol. exp. Morph. 18, 67-77.
GOLDING, D. W. (19676). Regeneration and growth control in Nereis. II. An axial gradient
in growth potentiality. /. Embryol. exp. Morph. 18, 79-80.
GOLDING, D. W. (1967 C). Neurosecretion and regeneration in Nereis. I. Regeneration and
the role of the supra-oesophageal ganglion. Gen. comp. Endocr. 8, 348-355.
GOLDING, D. W. (1967d). Neurosecretion and regeneration in Nereis. II. Prolonged secretory
activity of the supra-oesophageal ganglion. Gen. comp. Endocr. 8, 356-367.
GOLDING, D. W. (1967e). Endocrinology, regeneration and maturation in Nereis. Biol. Bull.
mar. biol. Lab., Woods Hole 133, 567-577.
GOLDING, D. W. (1974). Regeneration and growth control in Nereis. III. Separation of
wound healing and segment regeneration by experimental endocrine manipulation.
/. Embryol. exp. Morph. 32, 99-109.
BOILLY,
Hormone influence on Nereis regeneration
131
HERLANT-MEEWIS, H. & NOKIN, A. (1963). Cicatrisation et premiers stades de regeneration
pygidiale chez Nereis diversicolor. Annls Soc. r. zool. Belg. 93, 137-154.
HILL, S. D. (1972). Caudal regeneration in the absence of a brain in two species of sedentary
polychaetes. J. Embryol. exp. Morph. 28, 667-680.
HOFMANN, D. K. (1966). Untersuchungen zur Regeneration des Hinterendes bei Platynereis
dumerilii (Aud. et Milne-Edw.). Zool. Jb. (Zool. und Physiol.) 72, 374-430.
PORCHET, M. & DURCHON, M. (1968). Influence de la maturite genitale sur la regeneration
posterieire chez P. cultrifera G. C. r. hebd. Seanc. Acad. Sci., Paris 267, 194-196.
SCULLY, U. (1964). Factors influencing the secretion of regeneration promoting hormone in
Nereis diversicolor. Gen. comp. Endocr. 4, 91-98.
{Received 20 September 1973, revised 8 January 1974)
9-2